Bridging the gap between science and practice continues as a major problem in the effective delivery of treatments for alcohol abuse and dependence. Twelve Step Facilitation, an empirically supported therapy, is well positioned to bridge this gap. That is, although TSF is a highly structured manual driven approach, its basic tenets are familiar to many counselors and hence can be an effective 'bridge' to other empirically-supported, manualized approaches. In a randomized training study, we have demonstrated that intensive didactic training by ABR staff (Dr. Kathleen Carroll and Stu Baker, two of the original writers of the NIAAA Project MATCH TSF manual) dramatically increases 'real world' substance abuse counselors' ability to implement TSF effectively, compared to simply reading the manual. However, standard training models, regardless of their effectiveness, are not feasible to train large numbers of clinicians. Computer-assisted training models, which offer greater flexibility in the location and timing of training, as well as a high level of interaction, the opportunity to view effective implementation of the treatment, and numerous other advantages, have not been explored as a strategy to train practitioners and their ability to implement empirically validated therapies. In Phase I of the proposed project, we developed a single introductory module of computer-based training program in TSF ("CBT for TSF") and then conducted a randomized pilot study evaluating its effectiveness compared with reading the manual among 25 substance abuse counselors. Even in this comparatively small sample, the prototype 'CBT for TSF' program was associated with statistically significant differences in the counselors' ability to implement TSF, as measured by independent ratings of adherence and skill, as well as their knowledge of TSF, with large effect sizes (d=1.0). Thus, having met the a priori criteria for determining success for Phase I, we are now proposing a Phase II project in which we will complete the full "CBT for TSF" program and conduct a randomized training trial with 90 counselors. Counselors will be randomly assigned to one of three training conditions: Manual Only, the "CBT for TSF" program, or three days of face-to-face didactic training in TSF. The primary outcome will be pre-training/post-training independent ratings of the counselors' level of adherence and competence in delivering TSF, as measured by a standardized videotaped role play exercise. [unreadable] [unreadable]